Question

我是新手，需要帮助。我正在为我的论文制作一个基于代理的模型。我现在所困扰的是我的人口中的生与死。我有一个data.frame与男性，女性，年龄，preg，不preg和活着或死。我需要做的是 * for n for people for FOR FOR EACH PERSON

使用随机数决定性别（例如，＆lt; 0.5 =男性）
决定年龄，画一个随机数0到80
将Alive设置为TRUE
PREGO-男性和生殖前或生殖后女性，设为假
对于育龄女性，掷骰子

我也试图将所有这些数据存储在数据框中，以便为模拟创建一个开始的poplulation。

就像我说的那样，我只是盯着编码而且我在语法等方面受到了限制任何帮助甚至可能有用的资源都非常感谢

Answer 1

我的回答，没有循环：

#choose sample size and initialize dataframe with sex and age
samplesize <- 5000
df <- data.frame(sex = round(runif(samplesize)),
                 age = round(runif(samplesize, 0, 80)))

# all are alive and pregnant
df$alive <- TRUE
df$prego <- TRUE

# men and old/young women are not pregnant
df$prego[(df$sex == 1) | (df$age > 60) | (df$age < 14)] <- FALSE

# only some of the potentially pregnant women are pregnant
df$prego[df$prego == TRUE] <- sample(c(T,F), sum(df$prego), replace = T)

# replace 1s and 0s with M and F
df$sex[df$sex == 1] <- "M"
df$sex[df$sex == 0] <- "F"

Answer 2

这就是我要做的事情：

library(data.table)

#starting variables
probMale <- .5
ageMin <- 0
ageMax <- 80
preProductiveAge <- 15
postProductiveAge <- 44
probPregnant <- .3
sampleSize <- 10000

listOut <- list()
for(i in 1:sampleSize){
  sex <- sample(c('male', 'female'), size = 1, prob = c(probMale, 1-probMale))
  age <- sample(0:80, size = 1)
  alive <- TRUE
  if(sex == 'male'){
    prego <- FALSE
  } else if(age >= postProductiveAge){
    prego <- FALSE
  } else if(age <= preProductiveAge){
    prego <- FALSE
  } else{
    prego <- sample(c(TRUE, FALSE), size = 1, prob = c(probPregnant, 1-probPregnant))
  }
  listOut[[i]] <- data.frame(sex = sex, age = age, alive = alive, prego = prego)
}

df <- rbindlist(listOut)

我保留了在循环外定义的变量，因此您可以轻松修改。

Answer 3

无条件变量最初在tibble()内定义。条件变量在随后的dplyr::mutate()子句中设置。

library(magrittr)
person_count          <- 20
range_fertile         <- c(20, 45)
possible_genders      <- c("male", "female")
possible_ages         <- 1:80
pregnant_probability  <- .14


tibble::tibble(
  gender          = sample(possible_genders , person_count, replace=T),
  age             = sample(possible_ages    , person_count, replace=T),
  alive           = TRUE
) %>% 
dplyr::mutate(
  is_fertile_age  = (gender=="female") & (range_fertile[1] <= age & age <= range_fertile[2]),
  is_pregnant     = (is_fertile_age & sample(c(T,F), person_count, prob = c(pregnant_probability, 1-pregnant_probability), replace=T))
)

填充数据框

3 个答案: